Add test for Mysqli\Result

[phansys]

Q	A
Type	improvement
Fixed issues	n/a

Summary

This test is required in order to confirm cases like Result::rowCount() returning -1.

Related to #5915, #4880, #4870.

To-Do

• Find a query or a command sequence capable to reproduce the -1 result.

[derrabus]

I'm not really sure what is the purpose of that test. To me, it looks like we're asserting internal behavior of the MySQLi extension.

[phansys]

IMO, the purpose of this test is to ensure the behavior of the Result class (more precisely, of the Result::rowCount() method) regarding the different behaviors it adopts based on the value present at mysqli_driver::$report_mode.
As in #5915 I'm proposing changes that may affect these behaviors, I think is better to have the current expectations covered first.

[github-actions]

There hasn't been any activity on this pull request in the past 90 days, so it has been marked as stale and it will be closed automatically if no further activity occurs in the next 7 days.
If you want to continue working on it, please leave a comment.

[phansys]

IMO, this is still relevant.

[SenseException]

So mysqli has the behaviour to have a rowcount of -1 when something went wrong and this is a first step to cover such a case for the future of letting $result->rowCount() return only 0 or a positive integer in the future? I'm not sure I understand why this is in a separate PR.

[phansys]

So mysqli has the behaviour to have a rowcount of -1 when something went wrong and this is a first step to cover such a case for the future of letting $result->rowCount() return only 0 or a positive integer in the future? I'm not sure I understand why this is in a separate PR.

This test is provided in a separate PR because the behavior was not previously covered, and I think this is relevant regardless the merge of #5915. See #5915 (comment).

[SenseException]

@derrabus This is the behavior that currently is with mysqli and it's returning whatever mysqli does. #5915 is going to break it, but for now these tests prevent a bc break before 4.0.x. WDYT?

[derrabus]

I'm still not comfortable with that test tbh. But I'm fine merging it after some changes.